Dynamic Data Restoration from Multiple Recovery Sites Implementing Synchronous Remote Mirroring

1. A computer-implemented method, comprising: detecting an outage at a production site; transferring input/output (I/O) functionality to a first recovery site; in response to determining that the outage at the production site has been resolved, receiving a first out-of-sync bitmap from the first recovery site; receiving a second out-of-sync bitmap from a second recovery site, wherein the second recovery site serves as a synchronous mirrored copy to the first recovery site; merging the first and second out-of-sync bitmaps into a merged out-of-sync bitmap; receiving performance data which corresponds to achievable throughput from each of the first and second recovery sites to the production site; using the performance data to divide the merged out-of-sync bitmap into two portions, wherein a size of a first of the two portions correlates to the achievable throughput from the first recovery site to the production site, wherein a size of a second of the two portions correlates to the achievable throughput from the second recovery site to the production site; sending a request to the first recovery site for data which corresponds to the first portion of the merged out-of-sync bitmap; sending a request to the second recovery site for data which corresponds to the second portion of the merged out-of-sync bitmap; and receiving the data which corresponds to the first and second portions of the merged out-of-sync bitmap respectively, in parallel.

2. The computer-implemented method of claim 1 , wherein the data which corresponds to the first and second portions of the merged out-of-sync bitmap respectively, is fully received in substantially a same amount of time.

3. The computer-implemented method of claim 1 , comprising: detecting that a difference between actual performance and the performance data used to divide the merged out-of-sync bitmap into the two portions has exceeded a predefined threshold; sending a request to the first recovery site for an updated version of the first out-of-sync bitmap; sending a request to the second recovery site for an updated version of the second out-of-sync bitmap; merging the updated first and second out-of-sync bitmaps into an updated merged out-of-sync bitmap; using the actual performance to divide the updated merged out-of-sync bitmap into two updated portions, wherein a size of a first of the two updated portions correlates to actual throughput experienced from the first recovery site to the production site, wherein a size of a second of the two updated portions correlates to actual throughput experienced from the second recovery site to the production site; sending a request to the first recovery site for data which corresponds to the first updated portion of the updated merged out-of-sync bitmap; sending a request to the second recovery site for data which corresponds to the second updated portion of the updated merged out-of-sync bitmap; and receiving the data which corresponds to the first and second updated portions of the updated merged out-of-sync bitmap respectively, in parallel.

4. The computer-implemented method of claim 3 , wherein the data which corresponds to the first and second updated portions of the updated merged out-of-sync bitmap respectively, are fully received in substantially a same amount of time.

5. The computer-implemented method of claim 3 , wherein a total amount of time spent receiving the data which corresponds to the first and second updated portions of the updated merged out-of-sync bitmap is reduced by using the actual performance to divide the updated merged out-of-sync bitmap into the two updated portions.

6. The computer-implemented method of claim 1 , wherein the performance data is selected from the group consisting of: data granularity, response time, distance between sites, a number of data paths between sites, and data throughput.

7. The computer-implemented method of claim 1 , comprising: sending a data loss bitmap to each of the first and second recovery sites, wherein the out-of-sync bitmaps received from the first and second recovery sites are formed using the data loss bitmap.

8. A computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions readable and/or executable by a processor to cause the processor to: detect, by the processor, an outage at a production site; transfer, by the processor, input/output (I/O) functionality to a first recovery site; in response to determining that the outage at the production site has been resolved, receive, by the processor, a first out-of-sync bitmap from the first recovery site; receive, by the processor, a second out-of-sync bitmap from a second recovery site, wherein the second recovery site serves as a synchronous mirrored copy to the first recovery site; merge, by the processor, the first and second out-of-sync bitmaps into a merged out-of-sync bitmap; receive, by the processor, performance data which corresponds to achievable throughput from each of the first and second recovery sites to the production site; use, by the processor, the performance data to divide the merged out-of-sync bitmap into two portions, wherein a size of a first of the two portions correlates to the achievable throughput from the first recovery site to the production site, wherein a size of a second of the two portions correlates to the achievable throughput from the second recovery site to the production site; send, by the processor, a request to the first recovery site for data which corresponds to the first portion of the merged out-of-sync bitmap; send, by the processor, a request to the second recovery site for data which corresponds to the second portion of the merged out-of-sync bitmap; and receive, by the processor, the data which corresponds to the first and second portions of the merged out-of-sync bitmap respectively, in parallel.

9. The computer program product of claim 8 , wherein the data which corresponds to the first and second portions of the merged out-of-sync bitmap respectively, is fully received in substantially a same amount of time.

10. The computer program product of claim 8 , the program instructions readable and/or executable by the processor to cause the processor to: detect, by the processor, that a difference between actual performance and the performance data used to divide the merged out-of-sync bitmap into the two portions has exceeded a predefined threshold; send, by the processor, a request to the first recovery site for an updated version of the first out-of-sync bitmap; send, by the processor, a request to the second recovery site for an updated version of the second out-of-sync bitmap; merge, by the processor, the updated first and second out-of-sync bitmaps into an updated merged out-of-sync bitmap; use, by the processor, the actual performance to divide the updated merged out-of-sync bitmap into two updated portions, wherein a size of a first of the two updated portions correlates to actual throughput experienced from the first recovery site to the production site, wherein a size of a second of the two updated portions correlates to actual throughput experienced from the second recovery site to the production site; send, by the processor, a request to the first recovery site for data which corresponds to the first updated portion of the updated merged out-of-sync bitmap; send, by the processor, a request to the second recovery site for data which corresponds to the second updated portion of the updated merged out-of-sync bitmap; and receive, by the processor, the data which corresponds to the first and second updated portions of the updated merged out-of-sync bitmap respectively, in parallel.

11. The computer program product of claim 10 , wherein the data which corresponds to the first and second updated portions of the updated merged out-of-sync bitmap respectively, are fully received in substantially a same amount of time.

12. The computer program product of claim 10 , wherein a total amount of time spent receiving the data which corresponds to the first and second updated portions of the updated merged out-of-sync bitmap is reduced by using the actual performance to divide the updated merged out-of-sync bitmap into the two updated portions.

13. The computer program product of claim 8 , wherein the performance data is selected from the group consisting of: data granularity, response time, distance between sites, a number of data paths between sites, and data throughput.

14. The computer program product of claim 8 , the program instructions readable and/or executable by the processor to cause the processor to: send, by the processor, a data loss bitmap to each of the first and second recovery sites, wherein the out-of-sync bitmaps received from the first and second recovery sites are formed using the data loss bitmap.

15. A system, comprising: a hardware processor; and logic integrated with the hardware processor, executable by the hardware processor, or integrated with and executable by the hardware processor, the logic being configured to: detect, by the hardware processor, an outage at a production site; transfer, by the hardware processor, input/output (I/O) functionality to a first recovery site; in response to determining that the outage at the production site has been resolved, receive, by the hardware processor, a first out-of-sync bitmap from the first recovery site; receive, by the hardware processor, a second out-of-sync bitmap from a second recovery site, wherein the second recovery site serves as a synchronous mirrored copy to the first recovery site; merge, by the hardware processor, the first and second out-of-sync bitmaps into a merged out-of-sync bitmap; receive, by the hardware processor, performance data which corresponds to achievable throughput from each of the first and second recovery sites to the production site; use, by the hardware processor, the performance data to divide the merged out-of-sync bitmap into two portions, wherein a size of a first of the two portions correlates to the achievable throughput from the first recovery site to the production site, wherein a size of a second of the two portions correlates to the achievable throughput from the second recovery site to the production site; send, by the hardware processor, a request to the first recovery site for data which corresponds to the first portion of the merged out-of-sync bitmap; send, by the hardware processor, a request to the second recovery site for data which corresponds to the second portion of the merged out-of-sync bitmap; and receive, by the hardware processor, the data which corresponds to the first and second portions of the merged out-of-sync bitmap respectively, in parallel.

16. The system of claim 15 , wherein the data which corresponds to the first and second portions of the merged out-of-sync bitmap respectively, is fully received in substantially a same amount of time.

17. The system of claim 15 , the logic being configured to: detect, by the hardware processor, that a difference between actual performance and the performance data used to divide the merged out-of-sync bitmap into the two portions has exceeded a predefined threshold; send, by the hardware processor, a request to the first recovery site for an updated version of the first out-of-sync bitmap; send, by the hardware processor, a request to the second recovery site for an updated version of the second out-of-sync bitmap; merge, by the hardware processor, the updated first and second out-of-sync bitmaps into an updated merged out-of-sync bitmap; use, by the hardware processor, the actual performance to divide the updated merged out-of-sync bitmap into two updated portions, wherein a size of a first of the two updated portions correlates to actual throughput experienced from the first recovery site to the production site, wherein a size of a second of the two updated portions correlates to actual throughput experienced from the second recovery site to the production site; send, by the hardware processor, a request to the first recovery site for data which corresponds to the first updated portion of the updated merged out-of-sync bitmap; send, by the hardware processor, a request to the second recovery site for data which corresponds to the second updated portion of the updated merged out-of-sync bitmap; and receive, by the hardware processor, the data which corresponds to the first and second updated portions of the updated merged out-of-sync bitmap respectively, in parallel.

18. The system of claim 17 , wherein the data which corresponds to the first and second updated portions of the updated merged out-of-sync bitmap respectively, are fully received in substantially a same amount of time.

19. The system of claim 17 , wherein a total amount of time spent receiving the data which corresponds to the first and second updated portions of the updated merged out-of-sync bitmap is reduced by using the actual performance to divide the updated merged out-of-sync bitmap into the two updated portions.

20. The system of claim 15 , the logic being configured to: send, by the hardware processor, a data loss bitmap to each of the first and second recovery sites, wherein the out-of-sync bitmaps received from the first and second recovery sites are formed using the data loss bitmap.